A pervasive and diverse impairment of affect, unknown in its pathogenesis, characterizes the post-psychotic phase of 40% to 70% of schizophrenic patients successfully maintained in the community on maintenance neuroleptics. This severe distress is increasingly viewed as being causally related to drug noncompliance, poor social adjustment and subsequent relapse, even for neuroleptic maintained patients. The supplemental use of euthymic drugs in schizophrenia has yielded variable results among actively psychotic inpatients, yet suggests the presence of "types" of responsive patients. Very few trials, however, include outpatients in the recovery period (in spite of the wide spread use of these compounds) and none of these has convincingly ruled out milder forms of akinesia or akathisia as the toxins underlying impaired affect. The systematic study of 200 RDC schizophrenic patients, free of serious psychotic features and persistently distressed for at least four months post discharge is proposed. In a sequential strategy: (1) the contribution of akinesia and akathisia to these impairments will first be examined; (2) patients remaining distressed and meeting behavioral criteria for activation and/or inhibition will be randomized to double-blind trials of lithium or desipramine respectively; (3) selected patients who remain refractory will, in turn, be randomly assigned to a double-blind trial of diazepam (anxiety/activated) or to the intermittent, single dose reception of dextroamphetamine (defect state/inhibited).